Forecasting chaotic dynamic using hybrid system

TL;DR

This paper presents a hybrid neural network approach to improve the prediction of chaotic systems by synchronizing simulated dynamics with actual measurements, addressing limitations of parameter estimation.

Contribution

The paper introduces a novel hybrid system combining neural networks with simulations to enhance chaotic system prediction accuracy under partial information.

Findings

Successful synchronization of simulated and real system dynamics.

Effective prediction of atmospheric-inspired chaotic systems.

Applicable to low-dimensional systems with real-world complexity.

Abstract

The literature is rich with studies, analyses, and examples on parameter estimation for describing the evolution of chaotic dynamical systems based on measurements, even when only partial information is available through observations. However, parameter estimation alone does not resolve prediction challenges, particularly when only a subset of variables is known or when parameters are estimated with significant uncertainty. In this paper, we introduce a hybrid system specifically designed to address this issue. The method involves training an artificial intelligent system to predict the dynamics of a measured system by combining a neural network with a simulated system. By training the neural network, it becomes possible to refine the model's predictions so that the simulated dynamics synchronize with the actual system dynamics. After a brief contextualization of the problem, we introduce the hybrid approach employed, describing the learning technique and testing the results on two chaotic systems inspired by atmospheric dynamics in measurement contexts. Although these systems are low-dimensional, they encompass all the fundamental characteristics and predictability challenges that can be observed in more complex real-world systems.